[ALREADY REPORTED]Excessive Lateral Tire Grip - Taming the Viper

[Deano87]

Hey All.

 

I love flying in bad weather, 0 wind and glass smooth is a yawnfest for me. I like crosswinds, turbulence, the works! In other words what I'm saying is: I'm weird

 

And because most of my missions usually have crosswinds in them I've had issues with the way the Viper handles in crosswinds (specially landings) since it was first released.

 

My main issue is the sudden and extreme roll that happens when you try and touch the aircraft down while crabbed..Which is the correct technique for the F-16. In a 25 knot crosswind (the limit for the aircraft) Even on the very lightest touchdown with full into wind roll applied at the moment of touchdown you can easily strike the downwind wingtip which then usually results in some 2 wheel action and then a rollover, explosion, death etc etc. It IS possible but it's far too much like a lottery for it to be realistic.

 

So I started thinking, what is the problem here, whats causing the issue, and the answer is pretty simple. The tires have way too much lateral grip, they just refuse to slide sideways across the runway, and that's whats causing all the havoc. If you watch any video of the real F-16 landing in a strong crosswind, you can see they land fully crabbed and they keep that crab going the whole time they are aerobraking. Yes this does annihilate the surface of the tyres, but thats not the concern of the pilot at the time :D. What this shows is that real tyres have some amount of slip angle to them, which is not surprising considering the forces involved. If you watch a slow motion of a DCS crosswind landing with the current grip settings you can see the moment they make contact with the runway they grip hard and try and set off in the direction they are pointing, which happens to be into wind. This is what causes the unavoidable roll in the opposite direction, and inevitable wingtip strike etc etc.

 

So I decided to have a dig around in the config.lua file for the F-16 and see if I could edit the lateral tyre grip down to something more sensible. Indeed opening the file I found this:

 

As you can see both the main gear and nosewheel tires have a fixed lateral friction coefficient of 0.65.

 

I started editing this down to something that "felt" more realistic to me, Enough lateral grip to not be Tokyo Drifting everywhere but not too much that It felt like the aircraft had F1 spec qualifying slicks on it like it does now. I ended up with a figure of 0.40 for the main tyres and 0.42 for the nosewheel. These are numbers that just felt right to me, and I found them by trial and error.

 

You can see the results below in this video.

 

 

And here is a good real crosswind video I found from a similar angle, excuse the swearing.

 

 

Quite a difference as you can see! And it's not suddenly "easy", as far as I'm concerned it's gone from 50% chance of crashing to nicely challenging, which is as it should be. At 0.40 you still get a little roll on touch down but it's minor and because the tires can slide across the surface it actually manifests as a skipping or a bouncing from wheel to wheel which is exactly what the real aircraft does in this situation, I can hold the crab the whole way down the aerobrake as per the real aircraft and I'm just about running up against the limit on rudder to hold the nose where it needs to be, which makes sense considering 25 knots is the crosswind limit.

 

So its all well and good for me to come on here and say I think the tires are too sticky, how about some hard evidence that backs that up? I decided to do some digging and I found this:

 

https://ntrs.nasa.gov/archive/nasa/casi.ntrs.nasa.gov/19780005100.pdf

 

That is a NASA Paper from 1977 investigating Friction Characteristics of contemporary high performance jet fighter tires. Jackpot!

 

Here are some of the most revealing pages.

From those tests we can see that generally the lateral friction coefficient is FAR less than 0.65 as set in DCS and as speeds increase the friction coefficient decreases even further. The fastest they could test on their rig was 102 knots, which at 12 degrees slip angle gave a result of 0.44, So I think a setting of .40 in the sim is about spot on if you think that the F-16 is likely touching down around 140 knots.

 

Obviously as the sim uses a static number it means that as the aircraft slows the tyres are probably gripping less than they should, but this is less detrimental to the handling than being far too sticky on initial touchdown, IMHO anyway. I'll happily take the Viper sliding a bit more then it should at 50 knots instead of firing itself at the scenery at 140 knots.

 

I'm also pleased that the number I settled upon because it felt realistic to me is surprisingly close to the real figures.

 

It should also be noted that the aircraft is FAR easier to control during crosswind takeoffs with my modification as well, instead of trying lift the inside wheel it skips and slides sideways which is exactly what it should do.

 

Anyway thats my case for reducing the F-16 lateral tire grip.

 

I've attached my modified Config.lua below, If you want to try it out stick it into your root DCS files at DCS World OpenBetaModsaircraftF-16CFM, making sure that you backup the original version first!

 

Be aware that it will make you non Integrity Check compliant and that I made this in 2.5.5.41962, so may not work correctly for other versions.

 

Sorry for the long post.

 

Also in before somebody tells me its fine and I just need to "Git Gud Bro"

 

Cheers

D

 

---------------------------------------------------------------------------------------------

 

I've added a fun little 25 knot crosswind landing practice mission that you can try out to make a comparison with the normal grip setup. Its also just good practice as well to be honest, Enjoy.

 

---------------------------------------------------------------------------------------------

 

 

 

config.lua F-16 crosswind Landing 25kn.miz

Edited December 6, 2021 by Deano87
Adding real landing video.

[bbrz]

Wow, VERY nice mod!!!!

[b101uk]

you do know that many runways now (as opposed to in 1977) use significantly better asphalt compositions that result in notably higher friction coefficients in all conditions which can be as high as 0.75 in the dry with modern tyres.

 

 

 

 

this has been driven too by advances in tyre construction and composition and road safety and the fact that machines for testing skid resistance are much much more common now as are standards set in law for the minimum friction coefficients of roads dependent on if they are straight, a curve, leading into a braking zone or are on a gradient.

 

 

 

 

also a 25 knot crosswind limit on the aircraft dose dictate its is not that good in such cases, and I have to question why you are not taking out half of the crab just before touchdown.

[bbrz]

a 25 knot crosswind limit on the aircraft dose dictate its is not that good in such cases, and I have to question why you are not taking out half of the crab just before touchdown.

How do you know that 25kts is the actual aircraft limit without any margin etc.?

Why should you take out half the crab if the manual tells you to land with fully crabbed?

 

Btw, the higher friction coefficient would only apply in a 3 point attitude with full WOW.

Edited February 7, 2020 by bbrz

[Deano87]

The F-16 has been landed many a time in stronger cross winds then 25 knots, I’m the brother of somebody who has done it (operational reasons). But 25 knots is what the book says so that’s what I’m testing. As far as how I know it’s not realistic as is, because it doesn’t bare any relationship to how the real aircraft behaves in that situation, which my brother has also confirmed.

 

Please feel free to go try landing in a 25 knot direct crosswind with the stock settings and see how you get on. Hehe.

[ilikepie]

i wonder if this same thing causes the the "swing" when taking off that people have talked about?

[Deano87]

It’s certainly much more stable on takeoff with the lower lateral grip.

[randomTOTEN]

very very very difficult (impossible?) to land with a 25 knot crosswind. I also suspect this is related to other Viper ground handling issues.

[b101uk]

How do you know that 25kts is the actual aircraft limit without any margin etc.?

Why should you take out half the crab if the manual tells you to land with fully crabbed?

 

Btw, the higher friction coefficient would only apply in a 3 point attitude with full WOW.

 

 

 

 

he said that was the limit, I was just taking him at his word, and yes there will be a margin, but the fact it is a 25kts limit to start with dose tell its own story, as dose the 15kts gust limit.

 

 

 

 

because not every F-16 operating air force dose what the USAF dose, and for that matter not every USAF former F-16 pilots did what the USAF may require today, but the USAF can afford tyres and increased maintenance cost even though there is plenty of anecdotal evidence from former F-16 pilots that either will work but crabbing up to the flair then straightening out is smoother and less fuss.

 

 

 

 

if modern asphalt has a higher friction coefficient than 70's era asphalt, then it will be across the board, the same with modern tyre construction and composition.

[Deano87]

he said that was the limit, I was just taking him at his word, and yes there will be a margin, but the fact it is a 25kts limit to start with dose tell its own story, as dose the 15kts gust limit.

 

because not every F-16 operating air force dose what the USAF dose, and for that matter not every USAF former F-16 pilots did what the USAF may require today, but the USAF can afford tyres and increased maintenance cost even though there is plenty of anecdotal evidence from former F-16 pilots that either will work but crabbing up to the flair then straightening out is smoother and less fuss.

 

 

if modern asphalt has a higher friction coefficient than 70's era asphalt, then it will be across the board, the same with modern tyre construction and composition.

 

My brother was taught to land crabbed when he was flying the F-16 with the RDAF.

 

I don't doubt that its smoother to remove some of the crab, and will result in a smoother touchdown, but conversely I don't think that touching down at the published crosswind limit in the fashion they say to do in the book means that the aircraft should go out of control.

 

The crosswind limit of course has a buffer and the F-16 has been landed in up to 30 knots crosswind with no major controllability issues.

 

Clearly you have an issue with the numbers I've come up with, which is absolutely fine. But I'd really like to see a video of you landing in 25 knots + crosswind with the stock friction as for me its a monumentally tricky / almost impossible affair.

 

Also your friction numbers may well be correct for a modern tyre on a modern runway while stationary but as that Nasa paper shows as soon as you start moving and adding yaw angle the coefficient of friction starts decreasing. In the sim we have to work with a static number to try and represent a dynamic situation, I'd rather have that number set to be roughly right at the most critical moment which is the point of touchdown during maximum crosswind rather then it be spot on when I'm taxiing around at 20 knots.

 

And even IF the number is wrong, it's more about the end result, and to me with less grip the end result looks and feels much more convincing.

 

But thats just me.

 

[randomTOTEN]

Just because you found a single value that exerts control over this aspect of the flight dynamics doesn't tell the whole story. This might be the only value that needs to be changed, or there could be other values and files which affect this.

 

I've tried your values and they work fantastically well.

I would consider them a good temporary adjustment until we get an adjustment from ED.

Of course they fail the integrity check.

[Emmy]

I think the OP is very much on to something here!

[bbrz]

...the fact it is a 25kts limit to start with dose tell its own story, as dose the 15kts gust limit.

 

...there is plenty of anecdotal evidence from former F-16 pilots that either will work but crabbing up to the flair then straightening out is smoother and less fuss.

No, it doesn't tell any story. 25kts is a basically a standard 'limit' for most aircraft.

 

It's not about smoothness and pushing off the crab angle means definitely more, not less fuss.

Furthermore removing the crab at or prior the flare, can/will cause control difficulties due to the FBW logic.

Edited February 8, 2020 by bbrz

[Deano87]

Just attached a 25 knot crosswind practice landing mission to the original post for those that wanna try.

[b101uk]

@Deano87

 

 

its perfectly posable in a simulation to perturb a singular value of static friction to account for speed a yaw and produce a curve between the static value and a maximum angle and speed that will encompass most situations, and those numbers need not be user facing, they could be in other files or the .exe because they are universally applied to the plane specific values because they are constants.

 

 

 

 

To me you are only treating a symptom and not the cause, because there is definitely something else afoot that is the cause, because comparable values e.g. in the F/A-18 don't cause it to become unstable even when landing at much greater crab angles even if they are held to touchdown and part of the rollout, now granted the F/A-18 has a wider stance in the main gear and a longer travel and different damping, but IRL over-damping, too high an effective spring rate or to fast rebound or lack of rebound damping in combination can cause similar problem.

 

 

over damping can limit suspension compression under sudden loads but allows slower loads to use greater compression.

 

 

to fast rebound or no rebound damping allows the suspension to extend too fast after being compressed which can cause additional bounce as the waveform the wheel describes through relative vertical and forward travel when viewed from the side is a shorter frequency than that of the body it is ultimately attached via the suspension - think of some of the effects of a pogo stick.

 

 

And well spring rate along with overall travel have their own problems in conjunction with the above if they are to short, to high, not high enough or to long.

 

 

likewise physics in a 3D engine bring their own set of problems, in that the tick rate and the FPS can be largely divergent from one another, meaning you seldom visually see what happened in some short duration events that happened between frames, to you the observer a wheel at the moment of touchdown could look like it has moved slowly and naturally across a couple of frames but in physics it could have bounced once and already extending past the visual aspect, else one frame everything is normal and the next frame your doing 2000mph 5 miles away backwards because some force exceeded some computational limit thus produced a large value applied to the main body.

 

 

 

 

 

 

As for the NASA data, it is old and out of date both with respect to tyre construction and composition, not to mention the fact it was conducted not on actual runways but specially laid surface, the problem with which it hasn't had "traffic" on it and well anyone who has been around laying concrete and asphalt roads etc will tell you it takes a busy main road about a month for the surface to start providing maximum grip, because it takes time for the surface texture to develop and the stone chip to be properly exposed and profiled by the repeated action of tyres.

 

 

tyres themselves have come a long way since the 70's, it to the point today that the ply rating for a tyre is no longer indicative of the actual number of ply's in the construction but simply that of the load rating potential, tyres today themselves are lighter, suffer less centrifugal growth and stand up better to lateral forces as they deform less and are less susceptible to pressure change from heat changing the contact area, not to mention the fact that modern tyres have a far lower scrub-in period due to modern complex synthetic blends of rubber requiring less release agents from the moulds for the vulcanising process and don't require the preservative lacquer that natural rubber tyres do for storage, nor do they require as many in-service cycles before they reach their optimal chemical state of producing grip, not to mention the fact that the tyre carcass have more lives to be remoulded/re-treading which is common in aviation by the tyre's OM.

[Deano87]

I don’t disagree with any of your points. And the point I’m making is not that ED should necessarily take my numbers and run with them if there is some other underlying problem. But the currently behaviour of the aircraft isn’t realistic, and I’m showing what I’ve done to make it behave more realistically IMHO.

 

I presume you agree with me that the current behaviour isn’t correct?

[JTFF - Raph]

+ 1

[bbrz]

..the F/A-18 don't cause it to become unstable even when landing at much greater crab angles even if they are held to touchdown and part of the rollout...

Suggest to try and compare the F-16 to the very old F-15 module and you will notice that the F-15 is as unrealistic as the F-16 in crosswind conditions.

That's why I fear that this problem might not be solved at all from the ED side.

[BIGNEWY]

Thank you for the in depth analysis, I have added it to my report about crosswind landings for the team.

Edited February 9, 2020 by BIGNEWY

[Deano87]

Thank you for the in depth analysis, I have added it to my report about crosswind landings for the team.

 

Thanks BigNewy!

[b101uk]

I don’t disagree with any of your points. And the point I’m making is not that ED should necessarily take my numbers and run with them if there is some other underlying problem. But the currently behaviour of the aircraft isn’t realistic, and I’m showing what I’ve done to make it behave more realistically IMHO.

 

I presume you agree with me that the current behaviour isn’t correct?

 

 

 

I do agree that the currant behaviour isn't correct, but I am not convinced it is the problem, otherwise that singular values would cause more problems in other aircraft, some of which are definitely not known for their lateral grip.

 

 

something else that leads me to think the problem is elsewhere, is in your initial video, look at the wing tips, especially the right one and note the physics impulse as the right wheel touches the ground each time even in the second half of the video with your values, this leads me to think it is suspension values that show themselves when the FPM is a bit high and that there is a bit of a lack of flair, after all at the very minor FMP the RH wheel was touching the ground (after the initial touchdown) on the bounce you would not expect such an impulse.

[Deano87]

I do agree that the currant behaviour isn't correct, but I am not convinced it is the problem, otherwise that singular values would cause more problems in other aircraft, some of which are definitely not known for their lateral grip.

 

 

something else that leads me to think the problem is elsewhere, is in your initial video, look at the wing tips, especially the right one and note the physics impulse as the right wheel touches the ground each time even in the second half of the video with your values, this leads me to think it is suspension values that show themselves when the FPM is a bit high and that there is a bit of a lack of flair, after all at the very minor FMP the RH wheel was touching the ground (after the initial touchdown) on the bounce you would not expect such an impulse.

 

The wingflex is very much WIP and it wobbles around strangely even when flying (reaches max wing deflection at something like 5G), and the wingtips react in the incorrect direction for roll inputs, so I wouldn’t necessarily use that as a judge for anything that’s going on with the aircraft tbh.

 

Why not try watching a crosswind landing in super slow motion to see if the suspension is doing something weird? it should be possible to see once you’ve slowed the sim down to a small fraction of its normal speed, I’d be interested to see what you find,

 

Regarding that value causing problems in other aircraft, it does! The F-15C has the lateral friction coefficient set to 0.85 for all 3 tyres and it’s equally impossible to land in a 25 knot crosswind, let alone the 30 knot crosswind limit of the real aircraft. The real technique as described in the -1 for the F-15 is to land fully crabbed as per the F-16, which also doesn’t work and results in an instant tip-over. The F-15, like the F-16 is narrow track and shows this problem clearly. Editing the the F-15 grip down to lower levels also makes it much nicer to land. I’d expect that it has a similar effect with all other narrow track aircraft in the sim, but I’m going to do more testing with that.

[john9001]

I crab on final and straighten out to touch down.

[Deano87]

I crab on final and straighten out to touch down.

 

Have you tried that in my 25 knot crosswind mission in the first post? It won’t work hehe.

[john9001]

no, only in a Cessna 150 when I was taking flying lessons. it,s how the instructor said to do it.